{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "72f2cbf5",
   "metadata": {},
   "outputs": [],
   "source": [
    "import json\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "2ff4120b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/home/justas/projects/github/ProteinMPNN/outputs/example_pssm_outputs\r\n"
     ]
    }
   ],
   "source": [
    "!pwd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "d0600b01",
   "metadata": {},
   "outputs": [],
   "source": [
    "input_path = \"./parsed_pdbs.jsonl\"\n",
    "output_path = \"../../inputs/PSSM_inputs/\"\n",
    "mpnn_alphabet = 'ACDEFGHIKLMNPQRSTVWYX'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "71e81c08",
   "metadata": {},
   "outputs": [],
   "source": [
    "def softmax(x, T):\n",
    "    return np.exp(x/T)/np.sum(np.exp(x/T), -1, keepdims=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "ac59ccb3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "A 143\n",
      "B 143\n",
      "C 143\n",
      "A 227\n",
      "B 4\n",
      "C 227\n",
      "D 4\n",
      "E 227\n",
      "F 4\n"
     ]
    }
   ],
   "source": [
    "X_mask = np.concatenate([np.zeros([1,20]), np.ones([1,1])], -1)\n",
    "\n",
    "with open(input_path, 'r') as json_file:\n",
    "    json_list = list(json_file)\n",
    "\n",
    "my_dict = {}\n",
    "for json_str in json_list:\n",
    "    result = json.loads(json_str)\n",
    "    all_chain_list = [item[-1:] for item in list(result) if item[:9]=='seq_chain']\n",
    "    output_dict = {}\n",
    "    for chain in all_chain_list:\n",
    "        print(chain, len(result[f'seq_chain_{chain}']))\n",
    "        output_dict[chain+\"_coef\"] = np.ones(len(result[f'seq_chain_{chain}']), dtype=np.float32)\n",
    "        output_dict[chain+\"_odds\"] = np.ones((len(result[f'seq_chain_{chain}']),21))\n",
    "        random_bias = np.random.normal(size=(len(result[f'seq_chain_{chain}']),21)) #input your own PSSM bias\n",
    "        output_dict[chain+\"_bias\"] = softmax(random_bias-X_mask*1e8, 0.5) #making into probability distribution\n",
    "    np.savez(output_path+result['name'], **output_dict)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "ea3a4ed6",
   "metadata": {},
   "outputs": [],
   "source": [
    "a = np.load(output_path+'/3HTN.npz')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "63dcd23e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'PSSM probability')"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(a['A_bias'][10,:]);\n",
    "plt.xticks(np.arange(21), mpnn_alphabet);\n",
    "plt.grid(True)\n",
    "plt.ylabel('PSSM probability')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e34f2ce5",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}